Overview of CF and CFTR genotyping

Marci Sontag PhDAssistant Professor of Epidemiology

Colorado School of Public Health University of Colorado Denver, Aurora Colorado

and Children’s Hospital Colorado

Adapted from Welsh and Smith, Sci Am, 1995

Organ Dysfunction in CF

• Sinuses – Sinusitis, nasal polyps

• Lung – Endobronchitis, bronchiectasis

• Pancreas – Exocrine InsufficiencyCF Related Diabetes

• Intestine – Meconium ileusConstipation/DIOS

• Liver – Focal sclerosis

• Vas Deferens – failure to develop

• Sweat gland – salt-losing dehydration

22nd Annual North American Cystic Fibrosis Conference (NACFC) , 23 October 2008

• Genetic condition – 1/3,500 births; 35,000 individuals in US• Progressive lung disease

3

Cystic Fibrosis

• Median Predicted Survival: - 37 years

• Median Age at Death:- 26 years Patient Registry, Cystic Fibrosis Foundation, 2008, Bethesda

MD, USA (N=c.25,000)

4

Day 15 Week 8 Week 16 Week 24 Week 32 Week 40 Week 48-60

-55

-50

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

5

PlaceboVX-770

Cha

nge

in s

wea

t chl

orid

e co

ncen

trat

ion

mm

ol/L

(m

ean

, 95%

CI)

Hypothesis: Improving CFTR function will result in clinical benefit in patients with G551D

Treatment effect through Week 48

– 48.1 mmol/L P < 0.0001

Phase 3 Trial (Ramsey et al, NEJM, 2011)

First suggested: (Accurso et al, NEJM, 2010, N=39)Sweat Chloride

5

Day 15 Week 8 Week 16 Week 24 Week 32 Week 40 Week 48-5

0

5

10

15PlaceboVX-770

Abs

olut

e ch

ange

in %

pre

dict

ed F

EV

1(m

ean,

95%

CI)

Lung Function Improves with VX-770

Treatment effect through Week 48+ 10.5 %

P < 0.0001

Phase 3 Trial (Ramsey et al, NEJM, 2011)

Slide courtesy P. Farrell

Slide courtesy P. Farrell

Slide courtesy P. Farrell

Slide courtesy P. Farrell

Slide courtesy P. Farrell

Slide courtesy P. Farrell

Slide courtesy P. Farrell

By 2010, newborn screening was the most common diagnostic indication

U.S. CF Foundation Registry

0%10%20%30%40%50%60%70%80%90%

100%

Num

ber o

f Dia

gnos

es

OTHERMINBS

All new diagnoses reported to CFF in each year

Presented at NACFC, November 2011, Anaheim

Age of diagnosis has decreased with newborn screening

U.S. CF Foundation Registry

0.00.10.20.30.40.50.60.70.80.91.0

Med

ian

Age

Dia

gnos

is (y

ears

)

All new diagnoses reported to CFF in each year

<1 month of age

Presented at NACFC, November 2011, Anaheim

Complications in US • U.S. CF Foundation Patient Registry, 2000-2002• Comparison of

– Newborn Screening (NBS)– Symptomatic Diagnosis (SYMP)– Meconium Ileus (MI)– Prenatal

• Weight for age • Height for age• Hospitalizations• Pseudomonas aeruginosa infections

Accurso, Sontag, Wagener, J Pediatr 2005;147:S37-S41)

Newborn screened infants were less likely to be malnourished

(weight for age < 3rd percentile)

Accurso, Sontag, Wagener, J Pediatr 2005;147:S37-S41)

0

5

10

15

20

25

30

35

0-1 2-5 6-10 11-20

Age (years)

Perc

ent o

f pat

ient

s

NBSSYMPMIPrenatal*

*

**

** *

*

Accurso, Sontag, Wagener, J Pediatr 2005;147:S37-S41)

Children with CF who were newborn screened as infants fewer hospitalizations

0

10

20

30

40

50

60

0-1 2 - 5 6-10 11-20

Perc

ent o

f pat

ient

s ho

spita

lized

in y

ear

Age Group

NBSSYMPMIPrenatal

*

*

*

*

NBS, 985, 68%

Other, 243, 17%

MI, 215, 15%

Most infants under 2 years in 2010 were diagnosed early

U.S. CF Foundation Registry

• 83% of children < 2 years by the end of 2010 were identified by NBS or MI• The oldest baby in Texas identified under newborn screening was <1 in 2010

(~400,000 births/year, 60 babies with CF/year)

IRT/IRT

• Newborns receive 2 newborn screen tests– 1st before hospital discharge– 2nd at 2 week well baby check (mandated or extra sample

collected)• IRT is tested on both newborn screen blood spots• If both IRTs are elevated, child is recalled for a sweat

test (e.g. cutoffs at 100ng/ml and 70ng/ml)• No genetic testing is performed – no carriers are

identified

Introduction of mutation analysis to CF NBS

• CF Mutation identified in 1989• Wisconsin NBS program: 1991-92

introduced F508 (Gregg at al Am J Hum Genet 1993)

• Massachusetts: Multiplex CFTR Mutation Testing – 1999-2003 (Comeauat al Pediatrics)

(Riordan et al, Science, 1989)

IRT/DNA

• Newborns receive 1 newborn screen tests• IRT is tested on dried blood spot• If IRTs is elevated, same sample is tested for CFTR

mutations.• If 1 or more CFTR mutations are identified child is

recalled for a sweat test– 2 mutations – presumptive positive (sweat test)– 1 mutation – possible CF (sweat test)

Comparisons of Different Screens

IRT/IRT

IRT Cutoffs

Timing

Genetic Results

IRT/DNA

Tend to be 96-98 %

Earlier Diagnosis

Genetic Counseling Required

Tend to be >99th %

Must wait for 2nd test

No genetic info

IRT/IRT1↑/DNA

• Decrease 1st screen cutoff – 105ng/ml (99.7 %ile) to 97th %ile (~55ng/ml)

• Link 1st and 2nd screen specimens for each baby• Test 2nd screen ONLY if first screen > 97%ile• Mutation analysis if BOTH first and second

screen results > 97%

IRT/DNA-EGA• Newborns receive 1 newborn screen tests• IRT is tested on dried blood spots• If IRTs is elevated, same sample is tested for

CFTR mutations.– If 2 CFTR mutations are identified child is recalled for

a sweat test, presumptive positive– If 1 CFTR mutation is identified same blood spot tested

by expanded genetic analysis methods• If additional mutation(s) identified – sweat test• If no additional mutation identified – genetic counseling

• Fewer babies recalled for sweat tests

IRT/IRT has the highest sensitivity for the same cutoffs

However the positive predictive value is poor (many more sweat tests)Sontag et al, J Peds 2009

Goals for NBS Tests in CF• Minimize false negatives (Sensitivity)• Balance the number of false positives (PPV)• Provide a more specific diagnosis, i.e. DNA• Minimize the need for genetic counseling for

detection of carriers• Reduce parental stress

– Reduce the time to a diagnosis– Reduce the number of children/parents recalled for

testing• Reduce costs of screening and follow-up

Advantages to adding DNA testing to CF NBS

• Offers a more specific result in many cases – >60% of CF cases had 2 mutations.

• Can provide additional genetic information – Allow genetic counseling of parents of carriers

Challenges to adding DNA testing to CF NBS

• Clinicians ‘trust’ DNA– Need to educate clinicians that mistakes can

happen in all tests• Identification of carriers requires counseling• May miss individuals with rare mutations

(especially challenging in Hispanic populations in CF)

Selection of CFTR mutations• Only mutations known to cause CF should

be included in a panel• 23-mutation ACMG

– High degree of sensitivity– All mutations known to cause disease (special

case R117H*)• Additional mutations added when needed

for population coverage for regional differences

CLSI. Newborn Screening for Cystic Fibrosis; Approved Guidelin. CLSI document I/LA35-A. Wayne PA: Clinical and Laboratory Standards Institute, 2011

Allele Frequencies of CFTR Mutations From the ACMG-23 Panel Reported in Cohorts Detected Through CF NBS

CA* (23) MA* (24) NY* (24) CO* (25) WI* (26)

N=70 N=112 N=108 N=317 N=21

F508 75.3 67.9 57.4 71.3 66.7G542X 6.2 1.3 3.2 3.8G551D 3.7 3.1 1.4 1.4W1282X 3.7 1.8 0.9 1.1 2.4621+1G>T 2.5 0.4 0.5 1.6R553X 2.5 0.4 0.9 1.83120+1G>A 1.2 0.5 2.4I507del 1.2 0.5 0.7G85E 1.2 1.8 0.9R1162X 1.2 0.5N1303K 1.2 2.2 0.5 1.12789+5G-A 0.4 3.2 2.43849+10kbC>T 3.7 0.9 0.5 2.4R334W 2.5 0.5R117H † 4.0 0.9 ‡

R347P 0.5 2.4* CA = California; MA = Massachusetts; NY = New York; CO = Colorado; WI = Wisconsin.† Detection of this allele trans to a disease-causing mutation was excluded from percentages reported by theseauthors, but would have been > 1%.‡ Not tested in this mutation panel.

CLSI. Newborn Screening for Cystic Fibrosis; Approved Guidelin. CLSI document I/LA35-A. Wayne PA: Clinical and Laboratory Standards Institute, 2011

Balance of sensitivity/PPV

• Sensitivity: as long as one mutation from an affected patient is on panel, infant will be referred for sweat testing

• PPV: With the inclusion of too many mutations, more carriers will be called back for sweat testing

Detection of CF Cases and Carriers at Different Levels of Mutation Panel Sensitivity

Theoretical Population of 1000 Newborns With High IRT Referred for DNA Testing

0

10

20

30

40

50

60

70

80

90

100

Carriers 32 36 39 42 45 49 52 55 58 62 62 63 640 Mutations 7 6 4.4 3.4 2.5 1.7 1.1 0.6 0.3 0.1 0.0 0.0 0.01 Mutation 14 14 13 13 12 10 9 7 5 3 2 2 12 Mutations 7 8 10 12 14 16 18 20 23 25 26 26 27

50 55 60 65 70 75 80 85 90 95 96 97 98

CLSI. Newborn Screening for Cystic Fibrosis; Approved Guidelin. CLSI document I/LA35-A. Wayne PA: Clinical and Laboratory Standards Institute, 2011

Methods used

• Most state labs that are doing multiple CF mutation detection are using:– Luminex based assay (all FDA approved)– Hologic Inplex assay (ACMG23 FDA

approved)

– ACMG 23 – ACMG 23 plus additional mutations.

Reporting of resultsIRT/DNA

Test Result Value to Report Action RequiredIRT – no secondtier

Normal IRT level CF screen normal

Mutation analysis No mutations IRT levelNo mutations detected

CF screen normal

Mutation analysis One mutation IRT level and mutation Sweat chloride testing

Mutation analysis Two mutations IRT level and mutations Call PCPSweat chloride testing

Reporting of resultsIRT/IRT/DNA

Test Result Value to Report Action Required

IRT – no secondtier

Normal IRT level CF screen normal

Mutation analysis No mutations IRT levelNo mutations detected

CF screen normal

Mutation analysis One mutation IRT level and mutation Sweat chloride testing

Mutation analysis Two mutations IRT level and mutations Call PCPSweat chloride testing